Gleam is a finetune of Muse Glimmer focused on improving roleplay and writing performance using DPO with an emphasis on prose quality and humanlike characters. The dataset consists of approximately 1/3rds writing and 2/3ds roleplay.
I recommend grabbing the Q4_K_M gguf and using koboldcpp.
Training was conducted on 1xH200 for 5 hours using humanline dpo w/ LoRA.
model-00001-of-00002.safetensors 50 GBmodel-00002-of-00002.safetensors 9.7 GBLoRA/adapter_model.safetensors 1.1 GBmeta-models/Muse-Glimmer-30B